Are All Deep Learning Architectures Alike for Point‐of‐Care Ultrasound?: Evidence From a Cardiac Image Classification Model Suggests Otherwise. (24th December 2019)

• Record Type: Journal Article
• Authors: Blaivas, Michael; Blaivas, Laura
• Is Part Of: Journal of ultrasound in medicine. Volume 39:Number 6(2020)
• Page Start: 1187
• Page End: 1194
• Publication Date: 2019-12-24
• Subjects: artificial intelligence -- deep learning -- echo -- emergency medicine -- emergency ultrasound -- point‐of‐care ultrasound; Ultrasonics in medicine -- Periodicals; Ultrasonics; Ultrasonography; Ultrasonics in medicine; Electronic journals; Periodicals; Periodicals; 616.07543
• DOI: 10.1002/jum.15206
• Languages: English
• ISSNs: 0278-4297

Abstract

Abstract : Objectives: Little is known about optimal deep learning (DL) approaches for point‐of‐care ultrasound (POCUS) applications. We compared 6 popular DL architectures for POCUS cardiac image classification to determine whether an optimal DL architecture exists for future DL algorithm development in POCUS. Methods: We trained 6 convolutional neural networks (CNNs) with a range of complexities and ages (AlexNet, VGG‐16, VGG‐19, ResNet50, DenseNet201, and Inception‐v4). Each CNN was trained by using images of 5 typical POCUS cardiac views. Images were extracted from 225 publicly available deidentified POCUS cardiac videos. A total of 750, 018 individual images were extracted, with 90% used for model training and 10% for cross‐validation. The training time and accuracy achieved were tracked. A real‐world test of the algorithms was performed on a set of 125 completely new cardiac images. Descriptive statistics, Pearson R values, and κ values were calculated for each CNN. Results: Accuracy ranged from 96% to 85.6% correct for the 6 CNNs. VGG‐16, one of the oldest and simplest CNNs, performed best at 96% correct with 232 minutes to train ( R = 0.97; κ = 0.95; P < .00001). The worst‐performing CNN was the newer DenseNet201, with 85.6% accuracy and 429 minutes to train ( R = 0.92; κ = 0.82; P < .00001). Conclusions: Six common image classification DL algorithms showed considerable variability in their accuracy and training time when trained and tested on identical data, suggesting that not all will perform optimally for POCUS DL applications. Contrary to well‐established accuracies for CNNs, more modern and deeper algorithms yielded poorer results.